An Expedited Palliative Radiation Protocol for Lytic or Proliferative Lesions of Appendicular Bone in Dogs

Introduction

Malignant tumors of the appendicular skeleton are common and affect predominantly large- to giant-breed dogs. The most common bone tumor recognized in the dog is osteosarcoma (OSA), with an estimated 85% of all skeletal malignancies being osteogenic in origin.¹ The radiographic appearance of OSA in a long bone includes cortical bone lysis and/or a proliferative sunburst appearance; loss of the normal trabecular pattern in the metaphases; tumor bone extension into the soft tissues; and soft-tissue swelling adjacent to the primary tumor. Tumor types of nonosteogenic origin are seen less often and include fibrosarcoma, chondrosarcoma, hemangiosarcoma, and, rarely, metastatic tumors from other sites.² Benign tumors of bone are extremely rare.

The primary complaint upon presentation of dogs with lytic and/or proliferative bone lesions characteristic of OSA is progressive pain or lameness. One goal of therapy is to palliate pain. Amputation of an affected limb is one way to palliate bone pain, and amputation is considered to be a part of the standard of care for OSA. With OSA, surgical removal of the primary tumor either via a limb-sparing procedure or limb amputation followed by chemotherapy with platinum-based drugs or doxorubicin provides the longest reported survival times (ranging from 268 to 540 days).^3–7 Amputation in some situations, however, is not appropriate. The presence of other orthopedic or neurological conditions rendering a dog incapable of ambulation on three legs or the presence of metastatic disease are often reasons for not pursuing amputation. When amputation is not an option, other means of palliative therapy can be offered.

The goal of palliative therapy is alleviation of pain associated with bone lysis/proliferation and restoration of good quality of life for the patient for as long as possible. Some forms of palliative therapy may slow progression of disease; however, palliative therapy is not expected to stop tumor progression or provide a cure. The most effective palliative treatment reported is coarse-fraction radiation, which is successful in decreasing pain in the affected limb without decreasing the patient’s quality of life. In general, coarse-fraction radiation therapy consists of a variable number of fractions ranging from one to four large fractions (8 to 10 Gy) of radiation, administered over various time periods (range 7 to 28 days).^8–14 All of the various radiation therapy protocols are effective in alleviating or decreasing bone pain from the primary tumor; however, a superior treatment protocol, radiation source, and fractionation scheme have yet to be defined.^8–14

A novel palliative radiation protocol using two treatments of 8-Gy fractions given over the course of 2 consecutive days has been used at Washington State University since 1997. It was developed to minimize the time an animal needs to spend away from home to receive therapy, yet still provide enough radiation to achieve prolonged pain relief. Fraction size primarily determines late effects, and dose over time determines acute effects; therefore, giving two doses of 8 Gy on consecutive days was theorized to be no different in toxicity than previously reported protocols, because fraction size was similar to previous reports, and total dose was low even if given in a shorter time period. A lower initial total dose of radiation also allowed for additional therapy (another two courses of radiation) to be given with the return of pain and progression of the disease. The purposes of this retrospective study were to evaluate the effectiveness of this radiation protocol in terms of onset and duration of pain relief and survival times, and to compare the results with other published protocols. Prognostic indicators were also examined to determine whether predictions could be made about a dog’s individual response to treatment.

Materials and Methods

The radiation therapy database at Washington State University (WSU) College of Veterinary Medicine was searched for all animals that underwent radiation therapy for treatment of lytic/proliferative appendicular bone lesions between August 2003 and August 2006. The time interval was limited to 36 months to increase the reliability of follow-up information obtained from clients. A total of 58 dogs were identified in this search.

Diagnosis of a lytic/proliferative bone lesion was made radiographically. A presumptive diagnosis of OSA was made based on signalment, location of lesion, a thorough physical examination, and radiographic appearance of an appendicular bone lesion as judged by a board-certified radiologist. Fungal osteomyelitis may mimic a primary bone tumor in its radiographic appearance, but it is unknown in the Pacific Northwest. Although fungal osteomyelitis is a differential diagnosis for the lytic/proliferative lesions of bone observed in the dogs of this study, it was ruled out unless the dog had a history of travel to geographical areas of potential exposure.

Radiographic evidence of an osteolytic or proliferative lesion was present in all cases. A histological or cytological diagnosis was available for only 11 dogs, and all were diagnosed with OSA. All 58 dogs in this study had thoracic radiographs, and 44 of the 58 dogs had a complete blood count and serum biochemical profile before initiation of therapy. Once dogs were identified, their medical records were reviewed to obtain information regarding signalment, clinical signs and presentation, duration of signs prior to treatment, and serum alkaline phosphatase (SAP) activity prior to treatment. Histories of drug therapy were noted and defined as treatment with a nonsteroidal antiinflammatory drug (NSAID); an opioid or a corticosteroid at least 1 week prior to, during, or after radiation therapy; or bisphosphonate administration after radiation therapy. Metastasis to any site prior to initiation of treatment was recorded. The time to onset of a positive response following radiation therapy, duration of pain relief after radiation therapy, side effects of radiation therapy, survival time postirradiation, and cause of death were also recorded.

All dogs in this study received at least one course of palliative radiation consisting of a total of 16 Gy given in two fractions of 8 Gy each on 2 consecutive days (day 0 and day 1). All dogs were treated with an SL15 Elekta Linear Accelerator^a using a standard treatment field of 10 × 20 cm. Fifty-seven dogs received 6 MV photons with parallel-opposed fields, and one dog received 15 MeV electrons in a single field. Treatment planning was completed manually, and in most cases the field size included the majority of the bone affected. If the field was smaller than two-thirds of the bone length, port films were made to ensure that the tumor and a minimum 3-cm margin were being treated. When possible, a strip of skin was spared from the field to help avoid lymphedema. For each treatment, each dog was anesthetized with Desflurane,^b and the average time of anesthesia and treatment was approximately 10 to 15 minutes.

In 57 of the 58 cases, referring veterinarians were able to provide accurate information on the date and cause of death. Only one dog was completely lost to follow-up. Owners of 47 of the 58 dogs in this study were reached and able to supply data that contributed to the results presented in this study. Clinical signs upon presentation were defined as lameness (weight bearing or nonweight bearing), a palpable mass on the affected limb that was noted as being painful or nonpainful, or a history of crying out in pain.

Response to radiation therapy was evaluated by owners. A single person interviewed all owners and used a standard set of questions. An owner report of diminished pain was further characterized as to whether any of the following were recognized: an increase in activity, a decrease in lameness, a change from nonweight-bearing to weight-bearing lameness, a decrease in pain on palpation, decreased vocalization, or a combination of the above depending on the clinical signs upon presentation. Time to onset of pain relief was recorded as either 2 days, 1 week, 2 weeks, 3 weeks, 1 month, >1 month, or never. Duration of pain relief was recorded as lasting at least 1 week, 2 weeks, 3 weeks, 1 month, 2 months, 3 months, 6 months, or >6 months.

Specific questions were also asked of dog owners regarding skin irritation, hair loss, change of hair color, or increased pain postirradiation. Lack of response to treatment was defined as no improvement following radiation therapy or worsening of clinical signs after treatment. The time to the onset of response to therapy was defined by the total number of days postirradiation until the owner identified one of the above responses. Duration of response to therapy was defined as the time from the onset of a positive response to radiation therapy to the return or worsening of the original clinical signs. Survival time was calculated from the date of the initial radiation therapy to the date of euthanasia or death. If the dog was still alive at the time of analysis, the date of October 1, 2006, was used to calculate survival times to that date.

Survival time and duration of pain response following radiation therapy were calculated using Kaplan-Meier life table analysis.^c Discrete variables were analyzed for prognostic significance using Log Rank Mantel Cox or Breslow-Gehan-Wilcoxen tests for significance. Variables analyzed included sex; weight-bearing/nonweight-bearing lameness; pain on palpation; presence of a detectable mass on palpation; location of the presumed tumor; metastatic disease present at the time of treatment; use of NSAIDs, corticosteroids, bisphosphonates, or opioids; and SAP activity prior to radiation therapy. Cox proportional hazards model was used to analyze for correlations between age and weight and between survival and duration of pain relief. Spearman’s rank correlations were used to assess confounding variables of age, weight, and location of the tumor. A P value <0.05 was considered significant for all calculations. Dogs still alive at the end of the study were censored from all survival calculations. Dogs that had amputations were censored from all calculations except those related to onset and duration of pain relief.

Results

The canine population in this study included four intact males, 33 castrated males, and 21 spayed females. Gender was not statistically predictive of response or survival. Age range was 1.5 to 13 years, with a median age of 9 years. Body weight ranged from 20 to 78 kg, with a median of 45.3 kg. Multiple breeds were represented in this study, including rottweiler (n=10), Irish wolfhound (n=6), golden retriever (n=5), rottweiler mix (n=4), Labrador retriever mix (n=4), Alaskan malamute (n=3), German shepherd mix (n=4), collie (n=2), Newfoundland (n=2), mastiff (n=2), Saint Bernard (n=2), Labrador retriever (n=2), and one each of the following breeds: Australian shepherd, Pit bull terrier, Brittany spaniel, boxer mix, Great Dane, Doberman pinscher, Great Pyrenees, fila, Airedale terrier, Alaskan malamute mix, Scottish deerhound, and German shorthaired pointer.

All but one dog in this study had lameness at the time of treatment. Of the 57 dogs with lameness, 49 (84%) were weight-bearing, and nine (16%) were nonweight-bearing. A swelling or palpable mass on the affected limb that localized to the region of the radiographic lesion was noted in 32 (55%) dogs, and 20 (34%) dogs had pain on palpation of the affected limb. A combination of two or more of the above clinical signs was noted in 40 (69%) dogs. The durations of clinical signs prior to radiation treatment were known for 55 dogs and ranged from 7 to 300 days, with a median of 28 days and a mean of 52.8 days (±9 days).

For 47 of the 58 dogs in this study, information was obtained directly from the owners regarding response to therapy and duration of pain relief. Forty-three (91%) of 47 dogs had a positive response to radiation therapy determined by some degree of pain relief and increased limb function. Four (8.5%) of the 47 dogs did not respond at all to radiation therapy and showed no noticeable relief or improvement. Three of these dogs were euthanized for reasons relating to the primary tumor at 13, 48, and 94 days postradiation. The fourth dog that did not experience pain relief had an amputation 18 days following irradiation and was still alive at the close of the study (430 days). Of the 43 dogs that responded positively to radiation therapy, the time to onset of pain relief ranged from 2 to 60 days, with a median of 2 days. Thirty-three (77%) of the 43 dogs that responded had a response in <1 week after treatment, but 10 (23%) dogs took ≥2 weeks to exhibit a positive response after treatment. Of the 43 dogs that responded positively to therapy, the duration of response ranged from 12 to 503 days, with a median of 67 days and a mean of 99±16 days.

Weight-bearing dogs at the time of presentation showed a trend toward longer duration of pain relief, with a median of 88 days compared to 41 days in nonweight-bearing dogs (P=0.0621). Dogs that were presented with a palpable enlargement of the affected limb at the time of treatment had a significantly longer duration of pain relief, with a median time of 96 days (mean 127±26 days) compared to a median time of 56 days (mean 61±11 days) in dogs with no mass noted (P=0.0292). Tumor location [Table 1] was found to be statistically significant for duration of pain response and survival time. Dogs with tumors of the distal radius had a median duration of pain relief of 118 days (mean 159±33 days; P=0.0076). Dogs with tumors located in the distal radius had a median survival time of 217 days (mean 268±38 days), which was significantly longer than survival for dogs with tumors located elsewhere (P=0.0144 when analyzed as six variables; P=0.005 when analyzed as two variables) [Figure 1]. A correlation was found between increased age and tumor location in the distal radius (P=0.0263).

Five dogs had radiographic lesions consistent with pulmonary metastasis at the time of presentation. Six additional dogs were known to have developed radiographic lesions consistent with pulmonary metastasis following the initiation of radiation therapy, but not all dogs were radiographed prior to euthanasia or death. No statistical difference was detected in duration of pain control or survival when metastasis was initially present.

A total of 53 (91%) dogs received at least one NSAID (carprofen,^d deracoxib,^e piroxicam,^f or meloxicam^g) at various times during the course of their disease. A total of 29 (50%) dogs received treatment with at least one opioid (tramadol HCl,^h codeine,ⁱ or hydrocodone), and three dogs received at least one dose of corticosteroids (prednisone) prior to presentation for radiation therapy. Four dogs received at least one dose of a bisphosphonate (alendronate^j or pamidronate^k) at various time intervals after radiation therapy. A combination of two or more of the above drug types was given in 32 (55%) of the 58 dogs in this study. The most common drug combination was an NSAID given in conjunction with an opioid (n=27). No statistical difference in duration of pain control or in survival time was noted with use of any drugs. Dogs that received treatment with opioids tended to have shorter survival times, with a median survival time of 110.0 days (mean 147±25 days), compared to dogs that did not receive opioid treatment (median survival time of 191 days; mean 223±30 days). Although a trend was noted, these data were not statistically significant (P=0.0525).

Total SAP activity prior to radiation therapy was available for 44 of the 58 dogs in this study. Of these, 27 dogs had SAP activity above the reference range for the Washington State Clinical Pathology Laboratory (>89 U/L). Median SAP activity for all dogs in the study was 104 U/L (range 20 to 9816 U/L), while the median value of SAP for dogs with elevated SAP activity was 155 U/L (range 101 to 9816 U/L). The median survival time for dogs with increased SAP activity was 153 days (mean 206±34 days). A median survival time of 166 days (mean 189±41 days) was noted for dogs with normal SAP activity prior to treatment. No statistical difference was noted in response to duration of pain relief or in survival time (P=0.5860) for animals with elevated or normal SAP activity.

Side effects of radiation therapy were mild. The most common side effect noted by owners was hair color change in the radiation field (n=16). Hair loss or thinning was also noted (n=8). No severe acute or late side effects were noted. Three dogs developed pathological fractures. Two fractures occurred early (48 and 78 days following treatment); however, one dog experienced a documented pathological fracture 5 months following irradiation in the time period where late effects of bone radiation might be seen. Also, several dogs were euthanized shortly after having an acute recurrence of lameness, and recheck radiographs were not made.

Five of the 58 dogs in this study were returned for a second treatment of radiation therapy (two fractions, 8 Gy each, total radiation dose of 32 Gy), and three dogs were returned for a second and third treatment (two fractions, 8 Gy each, total radiation dose of 48 Gy). In each case, a positive response was noted after each treatment, with a decrease in the duration of response to therapy noted in all but one case. Dogs that received two treatments had a shorter survival time (median 96 days; mean 155±62 days) than dogs that received three treatments (median 271 days; mean 287±46 days) or one treatment of radiation (median 133 days; mean 174±21 days). While the dogs that received two treatments of radiation therapy had a trend toward shorter survival times, this difference was not statistically significant (P=0.5158). No evidence of detectable metastasis was seen in any of the above dogs at the time of initial treatment; however, three of the eight dogs developed metastasis 6 months to 1 year after initiation of radiation therapy. One dog developed metastasis to bone (survival time 271 days), and two dogs developed lung metastasis (survival times were 375 days and 166 days).

A total of three dogs in this study had amputation of their affected leg following radiation therapy. Two of these dogs received initial pain relief after irradiation; however, in one dog the pain returned in <14 days, and the owner decided to have the limb amputated 45 days after radiation therapy. In this case, no adjunctive chemotherapy was administered, and the dog survived a total of 112 days following radiation therapy. In another case, the tumor was advanced and very large in size when palliative radiation was given. The dog became systemically ill and developed severe edema in the affected limb 2 months after treatment. The affected limb was amputated, but the dog died 10 days later from other complications (total survival time 79 days). One dog perceived by the owner to not have any pain relief underwent an amputation 18 days following the first dose of radiation therapy. This dog received adjunctive chemotherapy with a combination of cisplatin and doxorubicin and was still alive at the time of study completion (430 days).

The most common cause of death was euthanasia due to pain from the presumed primary tumor site that no longer was controllable, resulting in loss of quality of life (n=29). Six dogs died from pulmonary metastasis. Other causes of death, such as suspected gastrointestinal ulceration (n=7) and renal disease (n=2), were not directly related to the primary disease. No information on cause of death was available for three dogs in this study. Ten dogs were still living at the time of study completion. Survival data for 57 of the 58 dogs revealed a range of 13 to 510 days following treatment, with a median survival time of 136 days and a mean survival time of 179±19 days [Figure 2].

Discussion

Radiation therapy is a widely recognized and accepted means of palliating bone pain caused by cancer in both humans and animals.^{10,13,15–22} Palliative radiation should not be employed to provide long-term or definitive tumor control in the treatment of OSA or other primary bone tumors.¹ For animals in which amputation or limb-sparing surgery and chemotherapy are not an option (due to either owner preference or financial circumstances, the presence of metastatic disease, and/or severe neurological or orthopedic disease), palliative radiation therapy offers a reasonable treatment alternative. Radiation therapy for palliation of pain should be designed to employ the simplest fractionation schedule that gives maximum relief with minimum morbidity, owner time commitment, and cost.²⁰

This study demonstrates that palliative radiation therapy given in two 8-Gy fractions on consecutive days resulted in pain relief, improvement of limb function, and enhanced quality of life in 91% of the dogs. The onset of pain relief and improved limb function were achieved in a relatively short period of time. Also, the results of this study regarding duration of pain relief and survival times are comparable with data published from other studies in which more prolonged protocols were used [Table 2].^{8,10,12–14,23} Many treatment protocols have been investigated with the objective of defining a superior approach to palliative treatment of appendicular OSA. To the authors’ knowledge, no best protocol for this purpose has yet been identified.

In both the human and veterinary literature, debate is ongoing over the most effective palliative fractionation scheme, because both abbreviated and extended protocols have similar results.^{8,10,12–14,16,21,23} Also, the lowest dose of radiation that will produce positive clinical results has not been identified. The concepts of maximum pain control, maximum quality of life, and minimal stress have guided the logic behind many of the abbreviated protocols investigated to treat canine OSA as well as painful bony metastasis in humans.^{8,13,14,16,21,24,25}

In the human literature, evidence is growing that supports the use of a single 8-Gy fraction for treating painful metastasis in bone. Results of numerous randomized trials investigating the efficacy of a single treatment, or multiple treatments that achieve a higher total dose of radiation over time, have shown no clear benefit of one protocol over another.^16,21 It is interesting to note that, even though well-organized studies have indicated the effectiveness of a single dose of radiation for palliation of human bony metastasis, protracted protocols are still the most popular therapy.¹⁶

When developing the palliative radiation therapy protocol employed at WSU, the goal was ease of therapy with minimal stress to both the dog and owner without compromising standards of care. The literature indicates that shortened protocols using 8-Gy fractions are as effective in alleviating bone pain as protracted protocols with a higher total dose of radiation. Because two doses of 8 Gy given a week apart is an accepted protocol, two doses of 8 Gy given on consecutive days is no different radiobiologically, except for the possibility of causing increased acute side effects.

Acute side effects of radiation occur within 3 months after radiation therapy and involve rapidly dividing tissues. In this study, acute side effects were rarely reported and inconsequential. Owners reported no compromise in quality of life. The most common acute side effect (occurring in 24 of the 58 dogs) was slight hair loss limited to the treatment field. Given that the major determinant of severity of acute side effects of radiation is the total dose of radiation administered, the incidence and degree of acute side effects were no worse than those previously reported, as a total of only 16 Gy was administered (less than the amount administered in many other studies).

In general, late-occurring side effects are seen at >3 months following treatment and involve tissues that are slow to proliferate, such as bone. Fraction size is the primary determinant of late effects. Late-occurring side effects were not a problem in this study, but they could be evaluated in only 32 dogs surviving >90 days after treatment and in only 16 dogs surviving >6 months after treatment. The two pathological fractures that occurred within the first 3 months after radiation were likely in response to progression of disease or increased use of the leg. One dog sustained a bone fracture 5 months following radiation therapy, and this event may have resulted from late effects but also could have been related to tumor progression. Hair color change in the radiation field is also a late effect and was commonly seen. Late side effects have also been infrequently reported in other studies where similar fraction sizes were used.

The incidence or severity of side effects in dogs of this study did not increase for those that received more than one treatment. These results indicate that up to a total of 48 Gy in 8-Gy fractions over an extended period of time could provide pain relief with minimal complications, making additional therapy a possibility for animals that respond well to initial treatment. All dogs returning for follow-up therapy had responded to the initial radiation treatment, and each responded to subsequent treatment; however, the duration of pain relief decreased with additional treatment in every dog except one. Studies have shown that humans who achieved a complete response (i.e., were completely pain free) following initial treatment were more likely to respond positively to re-treatment than those who only achieved a partial response (were not completely pain free) following the initial treatment.²⁶ When re-treated, the duration of response was typically longer in patients who initially had a complete response than in patients who initially had a partial response.²⁶ These same responses may occur in the dogs treated in this study; however, assessing the degree of response following treatment in companion animals is difficult, as owners are the only ones to make the decision for treatment and provide treatment outcome information. One dog in this study had a longer pain-free interval following the second treatment than the first, but this was not the typical result seen following treatment with subsequent fractions in either this study or in other animal and human studies. Because only eight of the 58 dogs in the study population were returned for re-treatment, the population analyzed is perhaps not representative of a larger population.

Dogs that were nonweight-bearing at the time of treatment had a shorter duration of pain relief following radiation therapy; however, this was not statistically significant when compared to animals that were weight-bearing at the time of treatment. Nonweight-bearing dogs may have had more advanced and aggressive disease. Dogs that had an enlargement of the affected limb at the time of treatment had a significantly longer duration of pain relief than dogs without enlargement.

The distributions of the appendicular lesions in this study were consistent with those previously noted in the literature for OSA, with the distal radius and proximal humerus being the most commonly recognized locations.^1,27 In this study, the location of the bone lesion was significantly correlated to duration of pain relief and survival time. Dogs with bone lesions located in the distal radius had significantly longer periods of pain relief and survival time than dogs with bone lesions located elsewhere. These results are contrary to previous studies, which have indicated that dogs with bone lesions consistent with OSA located in the proximal humerus tend to have a longer duration of pain relief and an overall better response to palliative irradiation.¹⁴ One reason for this difference may be related to planning of the radiation fields. In the present study, the majority of the length of the bone was treated in every dog, whereas in other studies the treatment field varied from 2 to 3 cm beyond the visible tumor and in general did not include the entire length of the bone. By consistently treating the entire bone as was done in the current study, a larger number of tumorigenic cells (especially on the periphery of the tumor and healthy bone junction) were possibly killed during treatment, resulting in longer duration of pain relief in radial lesions as compared to other studies. It is also possible that the planning of the radiation field in the previous studies more adequately dosed tumors in the proximal humerus than the field used here, which could result in the better response rate for this location noted by Ramirez et al.¹⁴

Administering NSAIDs or opioids to help control bone lesion pain is commonly done before definitive therapy such as amputation, before palliative radiation, or upon recurrence of pain following treatment. In this study, dogs that were placed on opioids for pain control over the course of treatment tended to have a shorter duration of survival than dogs that did not receive opioids. Possibly the dogs that received opioids were in more pain and were presented with more advanced or aggressive disease and, therefore, did not respond as well to palliative radiation. This would also correlate with a trend toward poorer overall survival, because the cause of death in the majority of animals was euthanasia resulting from relapse of pain that could no longer be controlled with medication. Also, dogs that received an opioid in combination with palliative radiation may have experienced superior pain relief, allowing them to be more active on a bone that had decreased structural integrity and increased risk of fracture. If a pathological fracture occurred without detection, this would likely result in an increase in pain that would be harder to control with analgesics, resulting in the decision to euthanize. The two dogs with known early fractures were receiving opioid medications at the time of bone fracture.

Total SAP activity values have been reported to be prognostic indicators of overall response to treatment in animals with appendicular OSA. In past studies, animals with elevated total SAP activity and bone isoenzyme SAP activity had shorter survival times following amputation and chemotherapy than did animals with normal SAP activity prior to treatment. ^28,29 The results of this study did not demonstrate any prognostic value in SAP activities with regard to response to palliative radiation or survival times. Mueller et al evaluated SAP activity in dogs with OSA that received palliative radiation therapy and also found no association with survival times.¹³ Ehrhart et al reported that dogs with OSA having increased SAP activity prior to amputation and chemotherapy had a median survival time of 177 days versus 495 days for dogs that had normal SAP activity prior to surgery and chemotherapy.²⁸ Similar results were reported by Garzotto et al who found a median survival time of 5.5 months for dogs with elevated total SAP activity prior to surgery and chemotherapy as opposed to a median survival time of 12.5 months for dogs with normal total SAP activity.²⁹ The median survival time for dogs in this study having increased SAP activity and treated with palliative radiation therapy is similar to the reported median survival time for dogs with increased SAP activity treated with amputation and chemotherapy. Further studies would be valuable to help determine if the activity of SAP prior to treatment could be used to help advise owners whether to pursue palliative radiation therapy versus amputation and chemotherapy.

The limitations of this study include those encountered in any retrospective study. The results of this study rely on owner-supplied information and memory, which may not be completely accurate all the time. Because of this, the current study was limited to a 3-year interval to increase the accuracy of the information collected from each owner. Accurate follow-up was obtained on all dogs in this study from referring veterinarians, which contributed to what the authors believe are accurate survival data. Inability to contact all owners led to incomplete data collection related to onset and duration of pain relief. The most reliable way to evaluate these dogs after treatment would have been to require rechecks at WSU at set time points following treatment. However, owners choosing this protocol wanted fast, easy, and inexpensive therapy.

Another limitation of this study is the lack of definitive diagnoses of the osteolytic/osteoproductive lesions for most dogs. A bone lesion in the metaphysis of a long bone, case signalment, and compatible history are suggestive of a primary bone tumor, most likely OSA. Radiographs of the affected bone showing lytic and/or proliferative areas were strongly supportive of a diagnosis of OSA. Biopsy or cytology of the bone lesions is necessary to achieve a definitive diagnosis, but even these are not 100% accurate. Biopsy and histopathological diagnosis were not required for a dog to be treated with palliative radiation at WSU. This could mean that animals that responded better to palliative treatment may not have had primary OSAbut rather another type of tumor that responded better to radiation therapy or one that was more slowly progressive in nature. Afungal or bacterial osteomyelitis would not likely respond to radiation therapy, as was noted in most dogs in this study. The strengths of this study were that every animal received the same radiation therapy protocol, which standardized treatment, and only palliative radiation therapy with no chemotherapy was evaluated.

Conclusion

This study has shown that a palliative radiation therapy protocol of two fractions of 8 Gy on consecutive days was acceptable in alleviating pain and restoring quality of life to many dogs with lytic/proliferative lesions of bone presumed to be OSA. The total dose of radiation and fractionation schedule did not compromise survival time, duration of pain relief, or quality of life when compared with previous reports. A positive prognostic indicator for survival was a lytic/proliferative lesion located at the distal radius. Positive prognostic indicators for duration of pain relief include tumor at the distal radius, bearing weight at presentation, and the presence of a visible mass at the time of therapy. The results of this study revealed an effective, expedited palliative protocol that may have benefits for owners selecting radiation therapy for their dogs instead of pain medication alone. The protocol presented in this study not only accomplished the goal of palliation of pain, but it also eased the stress of therapy and minimized the time spent in the hospital for both the dogs and owners.